metadata
license: apache-2.0
language:
- en
- zh
tags:
- unsloth
- QiMing
- vllm
- sales
- b2b
- Strategist
- saas
- fine-tuned
- instruction-following
- role-playing
- cognitive-simulator
- mlx
- mlx-my-repo
pipeline_tag: text-generation
model_name: QiMing-Strategist-20B
library_name: transformers
base_model: aifeifei798/QiMing-Strategist-20B-MXFP4
Wwayu/QiMing-Strategist-20B-MXFP4-mlx-4Bit
The Model Wwayu/QiMing-Strategist-20B-MXFP4-mlx-4Bit was converted to MLX format from aifeifei798/QiMing-Strategist-20B-MXFP4 using mlx-lm version 0.26.4.
Use with mlx
pip install mlx-lm
from mlx_lm import load, generate
model, tokenizer = load("Wwayu/QiMing-Strategist-20B-MXFP4-mlx-4Bit")
prompt="hello"
if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
messages, tokenize=False, add_generation_prompt=True
)
response = generate(model, tokenizer, prompt=prompt, verbose=True)